Turtle Trading Explained

Turtle Trading Explained

"I always say that you could publish my rules in a newspaper and no one
would follow them.The key is consistency and discipline.Almost anybody
can make up a set of rules 80% as good as what we taught our people.What
they couldn't do is give them the confidence to stick to those rules
even when things are going bad." Richard Dennis, Father of the
Turtles.Mid 1983. Famous commodities (futures) speculator Richard Dennis
argues with his buddy Bill Eckhardt about whether great traders can be
trained, or whether it is an innate ability. To settle the argument of
nature versus nurture, they decide to try and teach 13 beginners to
trade, and if they can master the rules, fund them with large trading
accounts. These beginners, whittled down from over 1000 applicants, are
known as the 'Turtles'. Over the next four years, the Turtles earned a
collective compound rate of return of over 80%. Argument settled.

Turtle Trading - What Markets?
As they were trading million dollar-plus accounts, the Turtles had to
choose liquid instruments, including US Treasury Bonds, Coffee, Cocoa,
Sugar, Cotton, Gold SIlver, Coper, major currencies and Indices, and
Oils.

Turtle Trading - Position Sizing
All good trading systems pay strict attention to position sizing. The
control of risk is the elementary building block upon which good trading
is based, and Turtle Trading is no exception. The Turtles used
'Volatility normalization' - a fancy way of saying that the more
volatile an instrument, the smaller the trade, meaning that every
instrument would (hopefully) carry the same dollar risk.This is where
the much-talked-about 'N' comes from.
'N' is the 20-day exponential moving average of the ATR (true range).
The formula to calculate 'N' is:-
(19 x Previous Days N + TR) / 20
Next step - figure out the 'Dollar Volatility Adjustment'. This is
simply (N x Dollars per point). This was done so the size of a 'Unit'
could be calculated. Each 'Unit' would account for 1% of the trader's
equity, in other words, a Unit equals:-
1% of account / Dollar Volatility

As an example, assume an instrument that moves $100 per point. Assume
also that the account size is $10,000. For simplicity, assume N = 0.1.
A Unit then , is:-
(1% x $10,000) / (0.1 X $100)
i.e. $10.

Turtle Trading - Risk Adjustment
Turtles had 'notional' sized accounts - although an account might
notionally start the year at $1,000,000, in the case of a loss of 10%,
the size of this account would be reduced by 20%. In other words the
trader would have to trade as if he only had $800K, not $900, until such
time as the account had got back to the starting figure.

Turtle Trading - Trade Entries
Turtles entered trades based on two systems, a 20 day breakout system,
and a 55 day breakout system. To use the first system, if the market
traded during the day or opened thru the 20 day high or low, that would
be a signal to enter.One Unit would be bought/sold to initiate the
position.If the previous signal would have resulted in a successful
trade, this signal would be ignored, in an attempt to avoid
'whipsawing'.
The second system fired signals if price exceeded the 55 day high or
low, and these signals were always taken, irrespective of previous
success/failure.

Turtle Trading - Adding to Positions
Once in a position, Turtles would add a Unit every 1/2 'N' advance, up
to the maximum number of units they were permitted (4 in a single
instrument, 6 in 'Closely Correlated' markets, such as Oil and Crude, 10
units in 'Loosely Correlated' markets, 12 units overall in a single
direction).
The prime directive in all of this was CONSISTENCY. As the majority of
trades failed, it was essential to be in on ALL of them, so as not to
miss the few huge winners that made the profits.

Turtle Trading - Stops and Exits
No trade was allowed to incur more than 2% of the account equity in risk
- in other words the Turtles used mental stops no more than 2 'N' away
from the position.
To exit from a System 1 trade, if the 10 day high (short trade) was
broken, that meant close the trade. Likewise if the 10 day low (long
trade) was broken, close th etrade. To exit from a System 2 trade, a 20
day breakout in the opposite direction would signal the end of the
trade.

Turtle Trading - does it work?
The Turtle system undoubtedly works. However, it requires iron willpower
to follow the rules, and not to try and 'bend' the mechanics of the
strategy. Most people are not mentally equipped to deal with the
constant losses, even though they are handsomely offset by the
occassional huge winner. If you have that kind of personality, perhaps
Turtle Trading is for you!

Comments:
Paul Cote
Turtle System from Russel Sands book Turtle Secrets, Tradewins: Buy the
highest high of last 20 days with stop at lowest low of last 10 days.
Sell the lowest low of last 20 days and stop out on highest high of
last 10. The system works pretty well. It certainly beats moving
average systems I believe. I have tried optimizing and of course you
can curve fit a number for every commodity, for every time, but I think
the 20/10 works well enough. This leaves you flat part of the time.
From Lebeau and Lucas, this is a vast improvement over the stop and
reverse systems.


Manfred
Newsgroups: misc.invest.futures
From: "Manfred"
Date: Thu, 1 Jul 2004 17:23:25 -0600
Local: Fri, Jul 2 2004 7:23 am
Subject: Before you pay TurtleTrader.com

Here is an excerpt from Curtis Faith, one of the original Turtles trained by
Richard Dennis, et. al., concerning Michael Covel, the sole operator of
TurtleTrader.com. Faith offers the trading rules for free that Covel at
TurtleTrader.com offers for $1000. See the site, www.originalturtles.org
for additional details, in addition to the quote below:

One of the sad realities of the trading industry and futures trading in
particular is that there are far more people making money selling systems
and "ways to make money trading" than there are people actually making money
by trading.

There are many "famous traders" who don't make money as traders. They make
money selling new trading systems, seminars, home study courses, etc. Many,
if not all, of these so called "experts" can't trade and don't trade the
systems that they sell.

Yes, this is also true of those selling the Turtle Trading Rules. Consider
the major sellers: the first, a web site, TurtleTrader.com, and the second,
a former turtle. Here's what they won't tell you:

TurtleTrader.com - A web site run mainly by one guy (an admittedly talented
web marketer that also has a pharmacy site and a site that sells personality
tests), TurtleTrader.com purports to have the actual Turtle Trading Rules
and sells them to you for about $1,000. The site is filled with huge amounts
of information about trading and bills itself as the "No. 1 Source for Trend
Following Worldwide".

What they don't tell you is that the site is run by a guy who doesn't even
trade his own rules - or trade at all for that matter - has never been a
successful trader, yet purports to be an expert on the "Turtle Trading
Rules" and trend following.

His rules, while reasonably close to the actual rules, differ in material
ways from those taught by Richard Dennis.

You won't get any expert advice from the guy who runs TurtleTrader.com. All
you will get is the regurgitation of advice from other traders not tempered
by the experience of a successful trading career. Paying for advice from
this source is a lot like hiring a blind guide.

In the final analysis, TurtleTrader.com is not any better than the other
scams and system selling hucksters he warns about. It is a site run by a guy
who appears to be more interested in taking his customers money than he is
in their success with the system he sells; someone who misrepresents himself
as an expert in trend following without mentioning that he doesn't trade.

Before dropping your money at TurtleTrader.com, carefully read what's
available for free at www.originalturtles.org

TURTLE TRADING IS NO SECRET

If there is a Holy Grail of trading (a colossal "if" in the minds of many), then it most likely consists less of trading systems and more of a set of processes, procedures, and precepts: Ride your winners and cut your losers short. When you are trading well, bet biggest, and when you are trading poorly, pull back or stop trading altogether. And last but not least, don't fight the trend.

While there are modifications and stylistic differences among traders as to the best way to execute this process - from scalpers who cut their winners short to top/bottom pickers who seek to ride trends from their origins (as opposed to profiting from what trend-followers refer to as "the meat in the middle") - few have fundamentally challenged these general trading concepts.

At the same time, few have embraced these general trading concepts as strongly as those who follow the Turtle trading methodology developed years ago by traders Richard Dennis and William Eckhardt. Although by now a legendary story in the annals of trading lore, the Turtle experiment is worth recalling for those who are unfamiliar with just what happened. Here, Richard Dennis explains the origins of the Turtle experiment in an interview with Jack Schwager in Schwager's excellent Market Wizards:

I have a partner who has been a friend since high school. We have had philosophical disagreements about everything you could imagine. One of these arguments was whether the skills of a successful trader could be reduced to a set of rules - that was my point of view - or whether there was something ineffable, mystical, subjective, or intuitive that made someone a good trader.

The resolution of this argument between Dennis and Eckhardt was the famous Turtle trading experiment in which 23 nontraders spent less than a month in training before being turned loose to trade based on the principles Dennis and Eckhardt prescribed. The results, according to Dennis, were average returns of 100% per year.

Too good to be true? Perhaps. But thanks to the efforts of subsequent generations of "Turtle traders," those who have been interested in, tantalized by, or just plain curious about how well the Turtle trading method works can find out for themselves.

SOME TURTLES ARE MORE EQUAL THAN OTHERS

All great successes beget both successors and imitators, and the Turtle trading methodology is no exception. Indeed, there are probably more websites and individuals trading off the Turtle trading reputation than any other single trading methodology. Fortunately, in TurtleTrader.com, aspiring traders have access to the real McCoy: the complete Turtle trading method as originally taught to Dennis and Eckhardt's first group of 23 aspiring traders. Not only does this website offer the Turtle trading course, but it also goes out of its way to educate visitors about the reason why trend-following - one of the mainstays of the Turtle trading method - is so successful, producing what they call "big picture, huge gains."

The educational sections of the website are valuable to anyone who trades - from beginners to veterans, and from stocks to commodities to Leaps? (long-term options). Nowhere else does a website take the time to explain what trend-following is and why it works - including examples of real-life, trend-following traders such as John W. Henry, whose results correlate strongly with the Turtle trading approach. More than a set of simple entry and exit rules, TurtleTrader.com also provides a great deal of information on the critical aspects of money management - from position sizing to handling profits and losses. Some have gone so far as to suggest that the success of the Turtle trading method is largely based on its unique money-management algorithms - perhaps even more than its strategies for getting into and out of markets. As such, the website features a section specifically devoted to the topic of money management.

While not giving away any secrets, TurtleTrader.com provides numerous articles on the statistical discoveries of real-life traders (and Bell Labs scientists) that support aggressive commitment to not just sound but strategic money management - a hallmark of Turtle trading.

Other sections include testimonials from trading houses like Galt Capital and Mulvaney Capital, as well as brief interviews with successful Turtle traders and clients like Tom Arnold of Plimsoll Capital and John Goodman of Pagnell Capital. Each of these profiles underscores the notion that "Turtle trading is brutal reality," as the website boasts. An even more entertaining testimonial is provided by TurtleTrader.com's section on, well, companies and institutions that (often unwittingly) found themselves on the opposite side of the Turtle trader - which means on the wrong side of the trend. If it is illuminating to read how a trader like Bill Dunn found great success trading with the trend, then it is equally fascinating to read how a Barings Bank or an Enron or a Metallgesellschaft found great failure in trading against it.

HYPE, HOPE, AND SECRETS

The bear market in stocks that began in 2000 provides an interesting context for all traders, and trend-followers are no exception. For all the methodologies that found such great success in the bull market, trend-following strategies are among the few that have been equally effective in the bear market. It is difficult to think of many other websites that sponsor a specific trading methodology and are as open about their methods, results, and returns as TurtleTrader.com. For example, the Market Proof section on the website provides data on trend-following gains in 2002, recent big Turtle trading wins, and examples of 10-year returns for various traders using either the Turtle trading approach or similar, correlated methods.

An entertaining aspect of TurtleTrader.com is the way it discusses and comments on the hype that surrounds trading and the world of Wall Street. Whether lambasting those who have pooh-poohed trend-following or pointing out the fallacies of traders for whom long-term luck has become a substitute for a winning market methodology, TurtleTrader.com serves a quasi-journalistic function in the world of contrarian financial commentary (I know, I know: How can trend-following ever be really contrarian?). Buy and hold, breakout chasing, the talking heads of CNBC ... all these baubles of the bull market just ended come in for regular skewering by the clear-eyed, take-no-prisoners mentality of TurtleTrader.com.

So what do you get for your money? The TurtleTrader.com website is free and includes an e-mail newsletter that will alert its subscribers to changes on the site, additions to the Turtle trader course, and discussions and commentaries on the world of trading. The Turtle trader course itself is $999 plus shipping - the same price as it was in 1996, when the course was first offered. The course includes a 200-page instruction text that covers all aspects of the Turtle Trading System, money management, an explanation of Richard Donchian's trading system (which supports the Turtle trading methodology), as well as information about client support, data on relevant hedge fund performance, offshore banking opportunities, and a guide to account management. In addition, the course provides an audio CD-ROM companion to the instruction text, and a data CD with historical volatility information for stocks and commodities. There is also a relatively new offering that consists of the same basic Turtle Trading System course, plus a module dealing with trading Leaps. This course is available for $1,699 plus shipping.

While TurtleTrader.com is obviously devoted to the Turtle trading method, what makes the method work is not the genius of the website developers, nor the genius of Richard Dennis or other traders identified with Turtle trading. In some ways, the secrecy to which many original Turtles were sworn has confused the issue of Turtle trading, making the methodology seem more clandestine and complex than it really is, while providing an unfortunate platform for less scrupulous promoters to offer the "secret" of Turtle trading to aspiring, yet unsuspecting, traders.

As one of the administrators of TurtleTrader.com suggested in e-mail, all Turtle trading amounts to, in the end, is trend-following. Indeed, there were successful trend-followers long before Richard Dennis and William Eckhardt's Turtle trading experiment, according to that administrator:

To some degree, we see the question as a continuation of the secrecy hype first revealed in Schwager's books. Great books, indeed, but the secrecy part of Turtles and trend following was misleadingÉ Given that so many other trend-followers were kicking butt at the same time of the Turtles' creation É

Turtle trend-following is not a secret. It simply takes proper teaching.

Trading Pairs

Trading Pairs
by Douglas S. Ehrman

Article Contributed by Active Trader Magazine

Pairs trading is a non-directional strategy that identifies two companies (or futures contracts) with similar characteristics whose price relationship is outside of its historical range. The strategy simply buys one instrument and sells the other in hopes that relationship moves back toward normal. The idea is the price relationship between two related instruments tends to fluctuate around its average in the short term, while remaining stable over the long term.

For example, the price relationship between Bank of America and U.S. Bancorp (BAC/USB) has remained relatively stable at a ratio of 1.5 over the past four years. During that period, its price ratio has been as high as 1.8 and as low as 1.35. The goal is to identify these stable relationships and buy the stock lagging the historical average and sell short the one leading it.

When trading pairs, you’re not concerned with how the individual stocks perform, but with their relative performance. As the market rises, both long and short stocks appreciate, so the pair’s overall value remains constant. Similarly, if the market declines, each stock drops. However, the pair profits as long as the long stock outperforms the short one.

Simply going long (or short) limits forecasting to a single direction, which means you can’t take advantage of the relative
values of all the instruments you track. Assume, for example, that you expect two stocks in the same industry to perform
much differently — one is expected to have a higher-than-average return and the other a lower-than-average return.

If you’re limited to going long, you’d buy the more attractive stock and ignore the one that’s expected to underperform.
In a pairs trade, however, you could enter a long position in the attractive stock and enter a short position in the less attractive one, which uses all available information and exploits the relative performance of both stocks.

2/20-day EMA breakout system

2/20-day EMA breakout system defined
by David S. Landry

Buy alert: If today's low and yesterday's low is greater than the 20-day EMA. This signal remains valid until the low touches or falls below the 20-day EMA.

Buy entry: Place a stop order 10 ticks above the two-day high. This will help ensure buying with the new trend and help to avoid false signals. Keep order until filled or as long as the buy alert is still valid.

Long exit: Place a stop equal to the 20-day EMA. Continue to update this stop daily to form a trailing stop.

See Graph


Sell alert: If today's high and yesterday's high is less than the 20-day EMA. This signal remains valid until the high touches or rises above the 20-day EMA.

Sell entry: Place a stop order 10 ticks below the two-day low. This will help ensure that you will sell with the new trend and help to avoid false signals. Keep order until filled or as long as the sell alert is still valid.

Short exit: Place a stop equal to the 20-day EMA. Continue to update this stop daily to form a trailing stop.

Channel-based Scalping

Moving average based trading systems still exist and work very well still, even today. Here is a basic scalping setup based on average range channel, which in turn is based on simple moving average, that works very well for many years.

A Channel Based Scalping Setup

Instead of jumping into the next topic of another usage combination based on simple moving average and average range, I will continue from where we left off in Part 2.

Here is the background info - emini S&P 45-minute bar time frame, 9 period simple moving average, average range channel offset by 2 times the average bar range. Not something very complex.

Isn’t scalping gets in and out of a position very quickly? With 45-minute bars, you sit on a position for at least 45 minutes. That is not scalping, isn’t it? Before you jump up and down complaining, just think - if you can get a significant bias at your side within the next 45 minutes trading S&P future, I think many of you will figure out your way to scalp within that 45-minutes.

My job here is to show you that even if you sit on a position based on the setup for the whole 45-minutes, you will still be profitable.

Here is the setup, we will go long if

1. The up side channel is broken

2. No matter it works out or not, we’ll exit by the end of the 45-minute bar

3. Being a person who has many things to do in life, I prefer trading in the morning only, so the setup will trade in the morning only

The complete version of the formula indicator Average Channel Scalper can be download here. You can also use the automatic Installer.

The actual code is very simple, just 6 lines.

makeindicator (avgbarrange, avgrange, data1, param1);
makeindicator (channel, avgrangechannel, data1, param1, param2);
$time_range := time >= maketime (9, 30, 0) and time < = maketime (12, 0, 0);
long_signal := $time_range > 0 and c >= channel.p2;
longatmarket (long_signal, defaultordersize, “Long”);
longexitnextclose (true, defaultordersize, “LExit”);

So how does the setup fair since 1998, here it is,

It works fairly well already, but not as consistent as we would like it to be. There is definitely room for further improvement on the raw system.

A Very Profitable Targeting Technique

Most beginners set position targets based on variation of the amount winning at the moment. For example, percentage of equity gained, percentage of gains to keep from the best level, or simply a fixed amount. That, will not help you capturing short and powerful moves.

For daytrading, time is your enemy (see article) thus you must take time into account when you are working on production trading systems. Within one 45-minute bar, what tools do we have to improve the performance? The answer is the average range of the bars. By cutting short the time to stay within the 45-minute bar using exit technique that close the postion early based on average bar range, we effectively lock in the possible profit within the bar and avoid staying on board where the possibility of a reversal is highly possible.

So, by adding an exit rule to get out based on the opening price of the 45-minute bar and a certain percentage away from that open, we have a new system.

The extra exit rule takes one extra line of code,

longexitnextopenoffsetlimit (true, avgbarrange * param3 / 100, defaultordersize, "LTarget");

Lets compare the following equity curve performance against the original one,

It performs a lot better than the raw system, isn’t it?

The system is not complex at all if you just take a closer look at the chart,

Summary

Do not underestimate the power of simple but statistically significant indicators. With properly constructed indicators, you will be able to find good trading setups everywhere. It takes time and patience to discover them.

I hope this example system will give you a head start.

Disucss this article.

1. It would be interesting to filter this system with volume considerations sort of in the style of market profile. High volume with breakout above/below the channel could produce effective results.

Buy-Open Sell-Close system

A Basic System - Buy Every Open and Sell At the Close

Here is a trading system based on 3 lines of code only, Buy Open Sell Close System. You can use the automatic Installer to install the indicator into your NeoTicker.

The system effectively consists of only 3 statements,

longatmarket (true, defaultordersize);
longexitnextclose (true, defaultordersize);
plot1 := currentequity;

What it does is very simple - it buys on every open and then exit at the close of the same bar. The performance of the system is reported through the plotting of the equity level.

I will explain the usage of each function in details.

LongAtMarket (condition, ordersize)

The LongAtMarket function instructs NeoTicker to place an order to buy at market with enough size (contracts or shares) to make the system net long at the ordersize, if the expression condition is evaluated to true (a value not equal to 0).

For example, if the trading system is flat at the moment, then LongAtMarket (true, 100) will buy 100 shares or contracts. On the other hand, if the system is currently short the market 200 shares or contracts, then LongAtMarket (true, 100) will buy 300 shares or contracts to make up the difference so that the system will become net long 100 shares or contracts when it is filled.

The condition parameter true is a constant which always evaluate to 1, that will ensure the LongAtMarket function to place an order every time it is evaluated.

Notice the function DefaultOrderSize used at the ordersize parameter is a trading system function that returns the user defined DefaultOrderSize which is set in the Indicator Setup window, or, through using the Symbol Info Manager.

LongExitNextClose (condition, ordersize)

When the parameter condition is evaluated to true (any value not equal to 0), an order will be placed to reduce the current long position by the size of the ordersize parameter at the close of the next bar. If the trading system is currently in a flat situation (no position at all) or in a short position, then even if the condition is evaluated to true, no order will be placed.

For example, if the system is net long 200 shares, then LongExitNextClose (true, 100) will place an order to sell 100 shares at the close of next bar. Or, if the system is net long 100 shares, then LongExitNextClose (true, 200) will place an order to sell 100 shares only, because the total long position is only 100 shares.

CurrentEquity

The CurrentEquity function is an account information function. It returns the current equity level of the trading system based on the accumulated performance of the system so far throughout the history of the chart.

For this simple system, we have already used 4 different trading system functions!

The System Performance

Here is how the chart looks like when the trading system is applied to daily emini S&P over the past 5 years.

One important issue related to trading system is the settings that you can adjust in the Indicator Setup window. For emini S&P, you need to set the price multiple to 50 and minimum tick size to 0.25. You can also change the Default Order Size right in the same system tab,

It is pretty obvious that this is a system that lose money. Well, the market is not really friendly towards somebody who simply try to take advantage of the classic market bias of going up over time, isn’t it?

A Twist In The System

Here is a new version of the same trading system with basic filtering, Buy Open Sell Close System Filtered. You can use the automatic Installer instead.

What this new system does is that it only trading on days that meet our criteria specified in the parameter.

makeindicator (cond, fml, data1, param1);
longatmarket (cond, defaultordersize);
longexitnextclose (cond, defaultordersize);
plot1 := currentequity;

Since I want the flexibility to change the condition easily for research purpose, the parameter is declared as a formula parameter, that enables the user the ability to access the Formula Editor when necessary. In our code, however, it is simply a string and in order to utilize it, we need to use the fml indicator to evaluate the formula stored within the parameter.

For example, if we specify the following condition to not trade on Tuesday and Friday,

The system will perform so much better (well, at least making some money) I think you will be shocked,

Summary

Remember that robust and profitable systems do not have to be very complex, its the ideas behind that counts. The usual obstacle for most people is that they do not know where to start and how to translate their ideas into something that can be backtested properly.

The subject of trading system writing, however, is very complex; due to the fact that there are so many types of trading systems based on vastly different underlying concepts. I will try my best to walk you through the basics techniques, and provide you with the necessary tools, so that you can translate your ideas into trading systems you want.

Open Range Breakout Daytrading System

Filed under: Trading System That Works — Lawrence Chan

Open range breakout system is a well known concept. It is a variation of the classic N-bar breakout system. This concept is so widely talked about because it works. I will show you a new twist of the concept that is never discussed anywhere else.

Open Range Breakout

What does the open range breakout system do?

1. it identifies the highest price and lowest price reached since open upto the Start Time,

2. long on a stop at the highest price and short on a stop at the lowest price obtained in #1,

3. only trade once per direction, so at most 1 long and 1 short positions taken per day,

4. always exit at the end time of the day,

5. when the range of previous trading session is above certain threshold of the average range of the previous trading days, no trade is taken for the day

Here is the trading system, Open Range Breakout. You can use Installer to install this system in your NeoTicker.

The system is written in formula and can be installed into NeoTicker by just copying that into the indicator directory.

The Performance

Once you have installed the indicator, you can now set up your chart.

The example I am going to use is 10-min ES, going all the way back to 1998. Remember to set the chart to the appropriate time range as the regular trading session of ES is 9:30 AM to 4:15 PM Eastern Time.

When you apply the system, remember to set your price multiple to 50 and min tick size to 0.25. The commission I used in my testing is 2.5 per trade.

Here is the result of the system.

http://newsletter.neoticker.com/wp-content/OpRngBO_fullday.gif

As a core system, with no bells and whistles, it works reasonably well.

One issue, though, is that the performance lately is not that good.

Our Twist

Go into the Chart Manager and change the chart’s trading time range from the original 9:30 AM - 4:15 PM, to 10:00 AM - 4:00 PM.

This technique is known as Data Reduction, which I have discussed in more details in another article titled “Daytrading the Emini” in the Technical Analysis of Stock & Commodities magazine, August 2003 issue.

Here is the performance of the system using reduced data.

http://newsletter.neoticker.com/wp-content/OpRngBO_10to4.gif

You probably do not believe it is the same system with the same default parameters, don’t you?

Conclusion

To improve a trading system does not necessarily involves a lot of parameter twistings. The more important thing to do is to understand why the system is not doing what it supposed to and find logic solutions to resolve the problem.

In this case, it is obvious that the opening 20 to 30 minutes is very chaotic and so is the last 15 minutes of trading. By removing those information, we improved the stability of our signals, thus the improvement in performance.

Discuss this article.

1.

There is a lengthy discussion and an ongoing realtime test of this kind of system at the Elitetrader Board: http://www.elitetrader.com/vb/showthread.php?s=&threadid=44092

As far as I remember they use 9:30 - 11:45 (EST) to get the range and some more conditions. Like do not trade on Thursdays(!) and do not trade when yesterday’s daily range was above its 5-day average.

Regards

Bernd Kuerbs

Comment by BKuerbs — July 1, 2005 @ 1:35 pm
2.

Open Range Breakout is a widely used concept in systematic daytrading. What we have here is a barebone system, by using data reduction, that evolved into a much improved model that has less drawdown and better overall performance. Everyone can use this as a starting point to further develop their own favour of this system. Not sure if the ideas in the Elitetrader thread can be incorporated into the system presented. It is a good exercise for our users.

Comment by Lawrence Chan — July 1, 2005 @ 4:39 pm
3.

[…] hen it is best to filter out days that are not likely to have expanded range. For example, The Open Range Breakout System will choose to not trade on days right after wide range days. General Solut […]

Intraday Moving Average System That Works

Formula 301 - #3 Intraday Moving Average Crossover with Position Sizing
Filed under: Formula 301 — Lawrence Chan

Many users are very interested in seeing example trading systems that utilize position sizing within NeoTicker. Personally I am extremely oppose to the idea of exposing beginners in trading system to the concept of position sizing as that is one of the most event dependent path leading to over optimization. Anyhow, here is a very good example showing you all what position sizing can do and cannot do.

Moving Average System Can Be Very Profitable Daytrading System

Most beginners in system design thought that moving average crossover does not work.

Here is a very simple moving average system that works.

1. Use Emini S&P 45 minute data series

2. Data range is 9:30 AM to 4:15 PM Eastern Time

3. Use NeoTicker’s moving average crossover system with moving averages set to 5 and 50 respectively

4. Single contract trading with $2.50 per side comission.

5. Most important of all, enable Close Position EOD in the Edit Indicator window, under System tab.

Here is the result.



Surprising, isn’t it?

The system is both stable and performing year after year. It is not something very complex either.

Position Sizing Based on Current Equity

Here is an indicator based on the same entry signal with position sizing.

I have not introduced any new trading system functions at all. As I mentioned in the previous tutorial, the basic trading system functions I have covered are enough for the design and implementation of many useful trading systems already.

The more interesting part of the code is the checking of current positions to see if you are over margin, if so, the system will do what your broker will do in real-life. That is, close out the part of the position that is over margin.

You can download the indicator here, Moving Average Crossover with Position Sizing. If you like, you can use this automatic Installer.

The main difference from the basic moving average crossover system is the use of currentequity as part of the calculation of the position size to be taken. This particular system takes a fixed amount as the basis per each contract to trade.

This method works best with system designed for trading with futures and other instruments based on fixed amount for margin.

I am going to show you a comparison chart of the system based on $5000 (5K System) and $9000 (9K System) risk per contract. Both systems will use the same Margin Requirement of $3000 per contract.

A few important points worth talking about here,

1. The 5K System does make 9 times the performance of the non-position sizing one.

2. The 5K System definitely performs much better than the 9K System one, but having the possibility of drawdown at 200K in a short period of time.

3. The 9K System performs way better than the reference equity of the original system.

4. We have not apply any money management rules to the trading system yet.

5. The 5K System has drawdown of about 50% of its equity multiple times. Do you think you can handle that?

The Overnight Risk

The basic moving average crossover system without position sizing actually performs better in terms of dollar value if traded by keeping overnight positions.

Then why do I emphasize that we are working on a daytrading system?

Here is a chart of the systems having their Close Position EOD option disabled.

Notice how bad it gets with the systems traded with position sizing.

Although I am always in favour of designing trading system with fixed position size, in this case, the position sizing version of the system exposed the weakness of the system when it is not traded as a day trading system.

We have a product called Equity Monaco that conduct this type of position sizing analysis. It works with NeoTicker’s system results, and text output from any other platforms. We offer this application for free because we believe that system traders should be aware of the risk of position sizing.

Why Does the System not Working Overnight?

The signal generated by the crossover is based on intraday data, thus, its directional bias has a shorter life span. When the unknown factor of overnight scenario changes is added to the equation, it is pretty obvious that the system will not be stable over time.

Summary

By applying position sizing rules to a trading system properly, you will be able to get spectacular returns for sure. That alone should not be the only reason you are trading a system at a particular size.

Position sizing is equity path dependent. In turns, the equity curve of your trading system is affected by extremely good (and bad) positions taken. Thus it is important to use position sizing correctly.

Formula 301 - Basic Consensus System

Formula 301 - #4 Basic Consensus System
Filed under: Trading System That Works, Formula 301 — Lawrence Chan

I am going to present a system that takes multiple technical indicators as part of its decision making. By combining these indicators through a simple scoring system, we get a very profitable daytrading system that works year after year.

(Lawrence - this system was designed many years ago (before year 2000) and was originally created for a trading platform that was very popular at the time. I was going through some old notes I have and adapted the code into NeoTicker. I am very glad that NeoTicker can implement the same system in 1/10th the amount of code required to implement the system. The original money management code has to be stripped due to various reasons, including NDA restrictions.)

Classic Multiple Signals System vs. Consensus Model

For many system designers, trading systems are signals, filters, and money management combinations. Especially for signals, they are very precise conditions from the crossover of 2 moving averages, the turning down of an oscillator, the breakout of a key price point, to the combination with specific trend quality. That is a pretty classical view of what trading models can be.

A more model view of trading system is that market direction changes due to many possible reasons, and if we can approximate the situation and estimate with high confidence of possible change in underlying condition, we are getting very good entry signals already.

Implementation Techniques

The main technique for implementing consensus model, is the use of scoring system. For each indicator or indicator group involved, we can assign scores to the group to identify the bullishness or bearishness of those indicators.

After scores are given for each indicator group, we can sum up the scores into a combined consensus score.

For simplicity sake, we can simply take long positions when the consensus score is bullish and go short when the score gets bearish.

The advantage of using a scoring system is that we can forget about the exact entry point signals and focus on the correctness in the directional bias, which is a way more important component that makes the system profitable.

The System

The system utilizes 3 completely different class of technical analysis to generate its score. In general, a +2 is assigned to the condition that is deemed as extremely bullish, +1 is assigned to the condition that is deemed as somewhat bullish, and vice versa. For indetermined condition, a score of 0 is assigned.

Interested readers can check out the indicators used within the script.

Here is the complete script.

One useful trick utilized in the script is the filtering of the signals generated at the end of a trading day. Such signals are not desirable because they force you to take on a trade at the open of the next trading day, which could be affected by the overnight events that altered the scenario completely.

The complete system is available here, Basic Consensus System. You can also use the automatic Installer.

You can now take a look at its performance.

As usual, the system is a daytrading system that always close its positions by end of day. It is applied onto emini S&P 45-minute data. The time range is the full regular trading hours from 9:30 AM to 4:15 PM Eastern Time. Commission is set at $2.50 per trade to reflect the cost.

The performance of this system is way better than any of its components.

Other Considerations

I call this system a basic consensus model because it is just a starting point.

Just think about the possibilities - we have not even utilized trend filtering, or even money management techniques. There are a lot of room for further improvement over this basic system.

You can also create your own consensus model using a completely different set of indicators and scoring rules.

End Notes

I do not expect this system to perform so well after it was dropped from real-time deployment due to its weak performance in year 1999. I think it deserves its spot in trading system history as it demonstrated the possibility of combining common indicators into a better trading model.

Lawrence

Discuss this article.

1.

If I built and tested this system correctly, it looks like it’s had flat performance since 12/1/2004 to 11/1/2004, on 57 trades. I guess this period is similar to the flat period in 1999?

Comment by Jay — November 5, 2005 @ 11:25 pm
2.

Jay, I assume you are talking about Nov 2004 to Jan 2005. One of the components used is Stochastics SlowK. The biggest enemy of oscillators is prolonged period of sideway market. The tight range behaviour across multiple days during this period caused those whipsaw trades. The interesting thing is, the drawdown during this period is still within the system’s normal volatility in its equity.

Comment by Lawrence Chan — November 6, 2005 @ 11:36 am
3.

Regarding your “end notes”. It’s hard to tell why performance was so weak in 1999. Would you explain further please.

Comment by Ed — November 18, 2005 @ 10:59 am
4.

The performance of the system in mid to end of 1999 is weak for a simple reason - the rules that define the stock market was broken. Serious investigations done after the melt down in year 2000 revealed many market participants, who traded during the end of the dot com era, are not following strict margin requirement, while their brokerages did not report or stop them from doing so. Thus prices was pushed around like unlimited amount of money is available. When strict margin requirement is enforced again, the system operates as expected.

Comment by Lawrence Chan — November 18, 2005 @ 3:59 pm
5.

Just out of curiousity do you optimize to arrive at the parameters you choose or is through chart overservation or perhaps someother perspective. Thanks for any comments you might have.

Comment by Ed — November 26, 2005 @ 3:53 pm
6.

No optimization applied. Most of the parameters are just common combinations. The scoring rules are also common interpretations of those indicators.

Stock Trading Using Swing Day Stochastics

Stock Trading Using Stochastics (End of Day)
Filed under: Cool Sets, Trading System That Works

Stochastics is one of the most widely used technical indicators in the world. Surprisely, most traders like to modify the original indicator into something else because they find the original implementation too noisy.

I am going to show you a trading setup using the original Stochastics SlowK indicator that works very well on daily data across many stocks that I think you will consider re-examine this classic technical indicator once more.

The Trading Setup

1. Applies on daily stock data

2. SlowK using the original 5, 3 parameter setting

3. This setup is long side only

4. Whenever SlowK crosses below the original oversold level, 25, we have a buy signal - go long at market

5. When SlowK recovers from the oversold level, and exceeded the original overbought level, 75, exit the open position at market

6. If within 5 days #5 has not happened, exit the open position at market anyway

Here is a chart showing the setup at work, Swing Stochastic Trading

Implementation

To implement the system, there are 2 issues that have to be addressed,

1. For long term testing of trading systems on stock data, it is inheritedly not stable because the surviving stocks we see today are in general long side biased as they survived for so long.

The solution is to include stock symbols that perform very badly for extended period of time in our testing so that we can tell if the system can survive such bad condition.

2. Position size cannot be fixed for stock trading system when we are testing over a long period of time because the stock price we see today in the chart is not really the price printed the day when the stock was traded due to split adjustment, dividend adjustment, etc. The price in the adjusted data series for symbols like IBM from 10 years ago is such a small number comparing to the current price level, it is making the trading system statistics we can collect less useful in general.

To validate the system we can check the system across a set of symbols to see if it performs in general pretty well over majority of the symbols, or, by using fixed size position, cross check each symbol over small partitions of the testing period to see if consistencies exists in the performance for each individual symbol over each individual time period, and as a whole.

I have written the system using formula indicator that takes into account of both position sizing and fixed position size (when the percent of equity parameter is set to 0).

The system is available here, SlowK Swing Trading System. You can also use the automatic Installer.

Performance

Some basic checking is done on the Dow Industrial Components.

I have set the system to start with $10K capital using cash only strategy. And always use round lot size of 100. Commission is taken into account too.

Here is the result for Exxon Mobile (XOM).

Remember that I mentioned including symbols with prolonged distressed perfomance? Here is the result for Intel (INTC).

Now, on American Express (AXP).

From these charts, we have learned a few things,

1. This sytem can handle prolonged bear market in a stock pretty well.

2. The system is pretty stable in terms of performance over time.

So, how does this system fair over all 30 Dow components, I ran the system through all of them in the pattern scanner, over the past 2000 trading days.

The system fair pretty well as there is only one symbol that is losing money.

You can download and install the group file contained the chart and the pattern scanner here,
EOD Stoch System Package. You can also use the automatic Installer.

End Notes

The most important thing that makes this system work is that it steps in when the situation looks very bad. It does not wait for, say, a confirmation day, before entering the market. That is the very essence of what SlowK is all about - It gives you a setup, so that you are prepared for what may happen next.

If you are a discretionary trader, think of what this setup means - if you are aware of the setup from daily data, you are then well prepared to take advantage of the possible turnaround in the market by monitoring the real-time action closely.

For system traders, the topic of integrating multiple time frames into the same system will be suitable for another full length article. In short, the risk taken by the daily system should be greatly reduced if real-time confirmation is taken into account.

Counter-Trend Trading

Here is an extremely simple trading system that violates the classic rule of trading, trend is your friend, with very good results. It is an excellent example that illustrates how counter-trend approaches work.

Simple Range Exhaustion System

1. Calculate the daily 20 period average range.

2. Today’s range must expand to exceed #1.

3. The 2-day combined range must expand to exceed 2.2 times of #1.

4. Go long at market if the market is trading in the lower portion of the 2-day range.

5. Protective stop is placed at 0.35 times #1 from the entry price.

6. If not stopped out, exit by 16:00 Eastern Time. i.e. a day trading system

To understand what the system does, just think about a rubber band that stretched thin. Once the force that hold up the stretched rubber band is gone, the rubber band will snap back to its normal self. Similarly, when the market moves in a particular direction too fast too soon, it will then snap back to a more sustainable level.

Almost all counter-trend trading techniques are based on similar principle. As oppose to wait for a confirmation before taking a position like trend followers, most counter-trend traders will try to time their entries as close to the extreme reversal points as possible to maximize the profits and minimize the risk exposures.

Oddball System

The original OddBall System no longer performs since end of year 2001 (see OddBall System - An Update). Here is an attempt to improve the system.

The Issue

The original Oddball system no longer performs due to the change in the behaviour in the NYSE advance issues has changed.

The change can be described in 2 different areas. First the range in which the advance issues data covered before end of 2001 is significantly smaller in comparison to the range we see in the past few years. Second, the extreme levels recorded in advance issues are significantly higher/lower now than the levels recorded before 2002.

Thus the original rate of change method than detects market turns in the early stage becomes more noisy and less reliable.

To improve the system, we can either use a finer resolution to obtain more details from the data, or, we can use some form of filtering to remove signals they are deemed to be less reliable. I am going to apply both techniques to see if we can improve the results.

A New Setup

1. S&P or emini S&P 30-Minute Chart, 9:30 AM to 4:00 PM ET

2. NYSE Advance Issues 30-Minute Chart

3. Use full trading day data only

4. Go long when rate of change on advance issues exceeded the prescribed long level

5. Go short when rate of change on advance issues exceeded the prescribed short level

6. Do not take the long signal in 4 when daily Stochastics exceeded the predefined filter level

7. Do not take the short signal in 5 when daily Stochastics dropped below the predefined filter level

How It Looks

Here is a chart with this system applied, (see diagram 1 above)

The Performance (see diagram 2 above)

The overall performance of this new variation of OddBall looks impressive, especially if we compare that to the significant drawdown in the original OddBall system.

End Notes

Although this new system does not suffer the huge drawdown as in the original version, nonetheless, its performance has dropped significantly comparing now to the period before 2002. One of the contributing fact is the contraction in daily trading range of the S&P. In order to further improve the system, we will need to take into account this factor.

To those who is interested in optimization, an optimized version of this new oddball system can do 30% better if I use long specific levels and short specific levels. The setting I am using now is symmetric for both long and short side.

Full article: Click Me